BCRP1;MDR1A/B;MRP2 COMBINATION KNOCKOUT MICE : ALTERED DISPOSITION OF THE DIETARY CARCINOGEN PHIP 2-AMINO-1-METHYL...

Page created by Perry Moore
 
CONTINUE READING
Bcrp1;Mdr1a/b;Mrp2 combination knockout mice : altered
disposition of the dietary carcinogen PhIP (2-amino-1-methyl-
6-phenylimidazo[4,5-b]pyridine) and its genotoxic metabolites
Citation for published version (APA):
Vlaming, M. A. J., Teunissen, S. F., Steeg, van de, E., Esch, van, A., Wagenaar, E., Brunsveld, L., Greef, de, T.
F. A., Rosing, H., Schellens, J. H., Beijnen, J. H., & Schinkel, A. H. (2014). Bcrp1;Mdr1a/b;Mrp2 combination
knockout mice : altered disposition of the dietary carcinogen PhIP (2-amino-1-methyl-6-phenylimidazo[4,5-
b]pyridine) and its genotoxic metabolites. Molecular Pharmacology, 85(3), 520-530.
https://doi.org/10.1124/mol.113.088823

DOI:
10.1124/mol.113.088823

Document status and date:
Published: 01/01/2014

Document Version:
Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:
• A submitted manuscript is the version of the article upon submission and before peer-review. There can be
important differences between the submitted version and the official published version of record. People
interested in the research are advised to contact the author for the final version of the publication, or visit the
DOI to the publisher's website.
• The final author version and the galley proof are versions of the publication after peer review.
• The final published version features the final layout of the paper including the volume, issue and page
numbers.
Link to publication

General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners
and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

       • Users may download and print one copy of any publication from the public portal for the purpose of private study or research.
       • You may not further distribute the material or use it for any profit-making activity or commercial gain
       • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please
follow below link for the End User Agreement:
www.tue.nl/taverne

Take down policy
If you believe that this document breaches copyright please contact us at:
openaccess@tue.nl
providing details and we will investigate your claim.

Download date: 23. Oct. 2021
Supplemental material to this article can be found at:
                                http://molpharm.aspetjournals.org/content/suppl/2013/12/13/mol.113.088823.DC1.html
1521-0111/85/3/520–530$25.00                                                                                                   http://dx.doi.org/10.1124/mol.113.088823
MOLECULAR PHARMACOLOGY                                                                                                         Mol Pharmacol 85:520–530, March 2014
Copyright ª 2014 by The American Society for Pharmacology and Experimental Therapeutics

Bcrp1;Mdr1a/b;Mrp2 Combination Knockout Mice: Altered
Disposition of the Dietary Carcinogen PhIP (2-Amino-1-Methyl-6-
Phenylimidazo[4,5-b]Pyridine) and Its Genotoxic Metabolites s

Maria L. H. Vlaming,1 Sebastiaan F. Teunissen,2 Evita van de Steeg,1 Anita van Esch,
Els Wagenaar, Luc Brunsveld, Tom F. A. de Greef, Hilde Rosing, Jan H. M. Schellens,
Jos H. Beijnen, and Alfred H. Schinkel
Divisions of Molecular Oncology (M.L.H.V., E.v.d.S., A.v.E., E.W., A.H.S.) and Clinical Pharmacology (J.H.M.S.), The Netherlands
Cancer Institute, Amsterdam, The Netherlands; Division of Pharmacy & Pharmacology, Slotervaart Hospital, Amsterdam, The

                                                                                                                                                                          Downloaded from molpharm.aspetjournals.org at ASPET Journals on February 20, 2015
Netherlands (S.F.T., H.R., J.H.B.); Laboratory of Chemical Biology, Department of Biomedical Engineering, Eindhoven University
of Technology, Eindhoven, The Netherlands (L.B., T.F.A.d.G.); and Department of Pharmaceutical Sciences, Science Faculty,
Utrecht University, Utrecht, The Netherlands (J.H.M.S., J.H.B.)
Received July 26, 2013; accepted December 13, 2013

ABSTRACT
The multidrug transporters breast cancer resistance protein                               reduced 41-fold in Bcrp1;Mdr1a/b;Mrp22/2 mice. Biliary and
(BCRP), multidrug-resistance protein 1 (MDR1), and multidrug-                             small intestine levels of PhIP metabolites were reduced in
resistance–associated protein (MRP) 2 and 3 eliminate toxic                               Bcrp1;Mrp2-deficient mice. Furthermore, in both knockout
compounds from tissues and the body and affect the                                        strains, kidney levels and urinary excretion of genotoxic PhIP-
pharmacokinetics of many drugs and other potentially toxic                                metabolites were significantly increased, suggesting that
compounds. The food-derived carcinogen PhIP (2-amino-1-                                   reduced biliary excretion of PhIP and PhIP metabolites leads
methyl-6-phenylimidazo[4,5-b]pyridine) is transported by                                  to increased urinary excretion of these metabolites and
BCRP, MDR1, and MRP2. To investigate the overlapping                                      increased systemic exposure. Bcrp1 and Mdr1a limited PhIP
functions of Bcrp1, Mdr1a/b, and Mrp2 in vivo, we generated                               brain accumulation. In Bcrp1;Mrp2;Mrp32/2 , but not Bcrp1;
Bcrp1;Mdr1a/b;Mrp22/2 mice, which are viable and fertile.                                 Mdr1a/b;Mrp2/2 mice, the carcinogenic metabolites N2-OH-
These mice, together with Bcrp1;Mrp2;Mrp32/2 mice, were                                   PhIP (2-hydroxyamino-1-methyl-6-phenylimidazo[4,5-b]pyri-
used to study the effects of the multidrug transporters on the                            dine) and PhIP-5-sulfate (a genotoxicity marker) accumulated
pharmacokinetics of PhIP and its metabolites. Thirty minutes                              in liver tissue, indicating that Mrp3 is involved in the sinusoidal
after oral or intravenous administration of PhIP (1 mg/kg), the                           secretion of these compounds. We conclude that Bcrp1,
PhIP levels in the small intestine were reduced 4- to 6-fold                              Mdr1a/b, Mrp2, and Mrp3 significantly affect tissue disposition
in Bcrp1;Mdr1a/b;Mrp22/ 2 and Bcrp1;Mrp2;Mrp32/2 mice                                     and biliary and fecal elimination of PhIP and its carcinogenic
compared with wild-type mice. Fecal excretion of PhIP was                                 metabolites and may affect PhIP-induced carcinogenesis as
reduced 8- to 20-fold in knockouts. Biliary PhIP excretion was                            a result.

   This work was funded by a grant of the Dutch Cancer Society [NKI 2003-
2940].
   1
                                                                                                                   Introduction
     Current affiliation: TNO, Zeist, The Netherlands.
   2
     Current affiliation: Netherlands Forensic Institute, Toxicology, The Hague,             The ATP-binding cassette (ABC) transporters P-glycoprotein
The Netherlands.                                                                          (P-gp or multidrug-resistance protein [MDR] 1 or ABCB1),
   Preliminary versions of this study were previously presented in the
following theses: Vlaming M (2009) ABC Transporter Compound Knockout                      breast cancer resistance protein (BCRP or ABCG2), multidrug-
Mice: Physiological and Pharmacological Characterization. Doctoral disserta-              resistance protein (MRP) 2 (ABCC2), and MRP3 (ABCC3) have
tion, University of Amsterdam, Amsterdam, The Netherlands (published
online at http://dare.uva.nl/document/135716); and Teunissen B (2011) Bio-                very broad and substantially overlapping substrate specificities
analysis and Metabolism of Tamoxifen and 2-Amino-1-Methyl-6-                              and can actively extrude potentially toxic compounds from
Phenylimidazo[4,5-b]Pyridine—Applications in Pharmacology. Doctoral
dissertation, University of Utrecht, Utrecht, The Netherlands.                            cells. They are expressed in organs with excretory functions,
   M.L.H.V. and S.F.T. contributed equally to this work.                                  such as liver, kidney, and small intestine, and are involved
   dx.doi.org/10.1124/mol.113.088823.
    s This article has supplemental material available at molpharm.                       in the elimination of endogenous and exogenous compounds
aspetjournals.org.                                                                        from the body. P-gp, BCRP, and, to a lesser extent, MRP2 are

ABBREVIATIONS: 5-OH-PhIP, 2-amino-1-methyl-6-phenylimidazo[4,5-b]-5-hydroxypyridine; ABC transporters, ATP-binding cassette trans-
porters; BCRP, breast cancer resistance protein; D3-PhIP, 2-amino-1-(trideuteromethyl)-6-phenylimidazo[4,5-b]pyridine; DMSO, dimethyl
sulfoxide; FVB, Friend Virus B type; IQ, 2-amino-3-methylimidazo[4,5-f]quinolone; LC-MS/MS, liquid chromatography–tandem mass spectrometry;
LLQ, lower limit of quantification; MDR1, multidrug-resistance protein 1; MRP2, multidrug-resistance–associated protein 2; MRP3, multidrug-
resistance-associated protein 3; N2-OH-PhIP, 2-hydroxyamino-1-methyl-6-phenylimidazo[4,5-b]pyridine; P-gp, P-glycoprotein; PhIP, 2-amino-1-
methyl-6-phenylimidazo[4,5-b]pyridine.

520
PhIP PK in Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;Mrp32/2 Mice                    521
additionally present in tissue sanctuaries, such as brain,          ABC transporters (Schinkel et al., 1997; Jonker et al., 2002;
testis, and placenta, where they protect these important            Vlaming et al., 2006, 2008, 2009a,b). These strains proved to
organs from entry of potentially toxic compounds. Whereas           be useful tools for pharmacokinetic studies. To be able to
P-gp, BCRP, and MRP2 are located at the apical membrane of          determine the combined effects of the apically located trans-
epithelial cells, transporting their substrates into bile, feces,   porters Bcrp1, Mdr1a/b, and Mrp2 on physiology and
and urine, MRP3 is located basolaterally in the intestine and       pharmacology, we have now generated Bcrp1;Mdr1a/b;
liver, where it transports its substrates into the blood            Mrp22/2 mice. Furthermore, we recently developed and
circulation (Borst and Elferink, 2002; Schinkel and Jonker,         validated a liquid chromatography–tandem mass spectrome-
2003; Leslie et al., 2005).                                         try (LC-MS/MS) assay for the quantitative determination of
   Besides endogenous compounds and a wide range of drugs,          PhIP and its metabolite N2-OH-PhIP in various matrices
the dietary heterocyclic amine carcinogens PhIP (2-amino-1-         from mice and expanded it to include multiple phase 1 and
methyl-6-phenylimidazo[4,5-b]pyridine) and IQ (2-amino-3-           phase 2 metabolites (Teunissen et al., 2010, 2011). We here
methylimidazo[4,5-f]quinoline) are also substrates for ABC          used the Bcrp1;Mdr1a/b;Mrp22/2 mice to investigate the
transporters in vitro and in vivo (Dietrich et al., 2001a,b; van    combined effect of Bcrp1, Mdr1a/b, and Mrp2 on the
Herwaarden et al., 2003, 2006; Leslie et al., 2005; Vlaming         elimination of PhIP and its metabolites in vivo. Furthermore,
et al., 2006; Enokizono et al., 2008). PhIP is the most             as the basolateral transporter Mrp3 is often upregulated
abundant heterocyclic amine in fried or cooked meat, chicken,       when Mrp2 is absent, and thus may compensate for the

                                                                                                                                                 Downloaded from molpharm.aspetjournals.org at ASPET Journals on February 20, 2015
and fish. In mice, PhIP primarily causes lymphomas and              decreased apical (biliary) efflux from the liver (Donner and
tumors of the small intestine, whereas in rats it causes colon,     Keppler, 2001; Zelcer et al., 2005; Vlaming et al., 2006), we
prostate, and mammary gland tumors. PhIP is rapidly taken           additionally investigated the pharmacokinetics of PhIP and
up after ingestion and heavily metabolized, leading to              its metabolites in Bcrp1;Mrp2;Mrp32/2 mice (Vlaming et al.,
detoxification products but also to activated, potentially          2009b).
carcinogenic metabolites (Supplemental Fig. 1; Gooderham
et al., 2001, 2002; Lauber et al., 2004; Nakagama et al., 2005).
Here, we focused on the precarcinogen N2-OH-PhIP (2-hydroxy-                           Materials and Methods
amino-1-methyl-6-phenylimidazo[4,5-b]pyridine), and the
                                                                       Animals. In compliance with Dutch legislation, mice were housed
derivatives of 5-OH-PhIP (2-amino-1-methyl-6-phenylimi-
                                                                    and handled according to institutional guidelines, and approval of the
dazo[4,5-b]-5-hydroxypyridine) and PhIP-5-sulfate, which            local animal care and use committee was obtained before the start of
represent breakdown products of the ultimate genotoxic              experiments. Bcrp1;Mdr1a/b;Mrp22/2 mice were generated by cross-
nitrenium radical cation that also forms mutagenic PhIP-            breeding Bcrp1;Mdr1a/b2/2 (Jonker et al., 2002) and Bcrp1;Mrp22/2
DNA adducts. Therefore, 5-OH-PhIP and PhIP-5-sulfate are            (Vlaming et al., 2009a) mice. The generation of Bcrp1;Mrp2;Mrp32/2
considered good markers for genotoxic exposure upon PhIP            mice was described before (Vlaming et al., 2009b). All animals were
exposure. PhIP-N-sulfate and N-acetyl-PhIP, the primary             of .99% Friend Virus B-type (FVB) background and 9–14 weeks old.
precursors of the genotoxic nitrenium radical, are too short-       Animals were kept in a temperature-controlled environment with a
lived to be detectable in pharmacokinetic studies. Further-         12-hour light/dark cycle. They received a standard diet (AM-II; Hope
more, part of the carcinogenic potential of PhIP may also be        Farms, Woerden, The Netherlands) and acidified water ad libitum.
                                                                       Chemicals. PhIP and its deuterated internal standard D3-PhIP
explained by the fact that PhIP possesses estrogenic activity
                                                                    [2-amino-1-(trideuteromethyl)-6-phenylimidazo[4,5-b]pyridine] were
and could therefore stimulate cell proliferation (Schut and         purchased from Toronto Research Chemicals (North York, ON,
Snyderwine, 1999; Frandsen and Alexander, 2000; Gooderham           Canada). N2-OH-PhIP was purchased from the National Cancer
et al., 2002; Lauber et al., 2004; Chen et al., 2007; Lauber and    Institute Chemical Carcinogen Reference Standard Repository at the
Gooderham, 2007, 2011).                                             Midwest Research Institute (Kansas City, MO). 5-OH-PhIP was
   Using Bcrp1- or Mrp2-deficient mice and rats, it was             a kind gift from Henrik Frandsen of the National Food Institute,
previously shown that Bcrp1 and Mrp2 significantly influence        Technical University of Denmark. Bovine serum albumin, dimethyl
the pharmacokinetics of [14C]PhIP in vivo (Dietrich et al.,         sulfoxide (DMSO), and formic acid were purchased from Merck
2001b; van Herwaarden et al., 2003, 2006; Vlaming et al.,           (Darmstadt, Germany) and were used as received. Methanol was
2006). However, since in several studies only radioactivity         obtained from Biosolve Ltd. (Amsterdam, The Netherlands). Distilled
                                                                    water was from B. Braun (Melsungen, Germany). Ketamine was from
was measured, no differentiation between parent PhIP and its
                                                                    Parke-Davis (Hoofddorp, The Netherlands), xylazine from Sigma-
mutagenic metabolites could be made. In Mrp2-deficient TR2          Aldrich (St. Louis, MO), and methoxyflurane (Metofane) from Medical
rats, it was found that Mrp2 influences the biliary excretion of    Developments Australia Pty. Ltd. (Springvale, VIC, Australia). All
PhIP, 49-OH-PhIP and some glucuronide conjugates (Dietrich          other chemicals and reagents were from Sigma-Aldrich (Steinheim,
et al., 2001a). Furthermore, in Bcrp12/2 mice, it was shown         Germany).
that Bcrp1 restricts the penetration of PhIP, N2-OH-PhIP,              Histologic, Clinical-Chemical, and Hematologic Analysis of
and 49-OH-PhIP into the brain or testis (Enokizono et al.,          Bcrp1;Mdr1a/b;Mrp22/2 Mice. Histologic analysis of male and
2008). Murine Mdr1a transports [14C]PhIP in vitro, but no           female mouse tissues (n 5 6), clinical chemistry analyses (including total
effect of murine Mdr1a/b on [14C]PhIP plasma elimination            and conjugated bilirubin, alkaline phosphatase, aspartate aminotrans-
was found (van Herwaarden et al., 2003), perhaps as a result        ferase, and alanine aminotransferase) on the serum of male and female
                                                                    mice (n 5 6), as well as standard hematologic analysis of male and female
of overlapping activity of Bcrp1 or Mrp2. The effect of other
                                                                    mice (n 5 6, twice within a time span of 1.5 years), were performed as
ABC transporters on the pharmacokinetics of PhIP and its            described (Vlaming et al., 2006).
metabolites has not been investigated yet.                             Real-Time Polymerase Chain Reaction Analysis. RNA iso-
   To investigate the overlapping or complementary roles of         lation, cDNA synthesis, and real-time quantitative polymerase chain
Bcrp1, Mdr1a/b, Mrp2, and Mrp3 in vivo, we generated a set of       reaction analysis on liver, kidney, and small intestine of male mice (n 5
compound knockout mice, deficient in up to three of these           3) were done as described (van Waterschoot et al., 2008).
522        Vlaming et al.

   Pharmacokinetic Experiments. For oral administration, 10 ml/g             Statistical Analysis. Unless otherwise indicated, the two-sided
of body weight of a 0.1 mg/ml PhIP solution in 20% (v/v) DMSO, and        unpaired Student’s t test was used to assess statistical significance
5% (w/v) D-glucose was dosed by gavage in the stomach of male mice.       of differences between two sets of data. For analysis of statistical
For intravenous administration, 5 ml/g of body weight of a 0.2 mg/ml      difference when compared with data below the detection limit, cal-
PhIP solution in 20% (v/v) DMSO and 0.9% (w/v) NaCl solution              culations were made based on the LLQ 6 LLQ. Results are presented
was injected into the tail vein of male mice. Animals were killed by      as the means 6 S.D. Differences were considered statistically significant
terminal bleeding through cardiac puncture under methoxyflurane           when P , 0.05.
anesthesia, and organs were removed. Intestinal contents and tissue
were separated.                                                                                         Results
   Biliary Excretion of PhIP and Its Metabolites. Gallbladder
cannulations in male mice were performed as described (van                   Macroscopic and Microscopic Analysis of Bcrp1;
Herwaarden et al., 2003). After cannulation, PhIP was administered        Mdr1a/b;Mrp22/2 Mice. Bcrp1;Mdr1a/b;Mrp22/2 mice were,
intravenously at 1 mg/kg as described above. Bile was collected in        like Bcrp1;Mrp2;Mrp32/2 mice (Vlaming et al., 2009b), viable
15-minute fractions for 60 minutes after administration of PhIP.          and fertile, and they had normal life spans, body weights, and
Subsequently, mice were killed by cardiac puncture and plasma and         anatomy. Adult male Bcrp1;Mdr1a/b;Mrp22/2 mice had a 46%
organs were collected.
                                                                          increased liver weight compared with wild-type mice (7.0% 6
   Fecal and Urinary Excretion of PhIP and Its Metabolites.
                                                                          0.5% versus 4.8% 6 0.2% of body weight, respectively, n 5 5; P 5
Male mice were individually housed in Tecniplast metabolic cages
                                                                          2.8 × 1025). As previously described, the liver weight of male

                                                                                                                                                      Downloaded from molpharm.aspetjournals.org at ASPET Journals on February 20, 2015
(Milan, Italy) and allowed 24 hours to adapt before PhIP (1 mg/kg)
was injected into the tail vein, as described already herein. Feces and   Bcrp1;Mrp2;Mrp32/2 mice was ∼70% higher than that of wild-
urine were collected over 24 hours. Subsequently, mice were killed by     type mice (Vlaming et al., 2009b). Despite the increased liver
terminal bleeding through cardiac puncture under methoxyflurane           size, detailed microscopic analysis of liver sections did not re-
anesthesia. Organs were removed, and intestinal contents (feces) and      veal obvious pathologic changes in either strain. An increase
tissue were separated.                                                    (somewhat smaller) in liver weight was previously seen in other
   Sample Preparation. After sampling, urine, bile, and plasma            Mrp2-deficient mouse strains (Vlaming et al., 2006, 2008, 2009a;
samples were snap-frozen. Tissue and feces samples were weighed           Tian et al., 2008).
before snap-freezing. The complete organ or the total volume of
                                                                             Plasma Clinical Chemistry and Hematologic Analysis
sampled feces was homogenized in a 4% (w/v) bovine serum albumin
                                                                          of Bcrp1;Mdr1a/b;Mrp22/2 Mice. Plasma clinical chemistry
solution using a Polytron blender. Feces homogenates were centri-
fuged for 10 minutes at 11,300g. A 10-ml volume of bile was diluted in    parameters in Bcrp1;Mdr1a/b;Mrp22/2 mice did not show
400 ml of 4% (w/v) bovine serume albumin solution.                        consistent significant differences from wild-type, except for
   Sample Processing. A 100-ml aliquot of plasma, diluted bile,           increased plasma bilirubin levels (conjugated and unconju-
tissue, or clear supernatant of centrifuged feces homogenate was          gated), as previously shown for Bcrp1;Mrp22/2 mice (Vlaming
processed immediately after thawing at ambient temperature by the         et al., 2009a). Additional deletion of Mdr1a/b did not further
addition of 300 ml of internal standard (D3-PhIP) solution in             affect the plasma bilirubin levels of the mice (data not
acetonitrile. The mixture was vortex-mixed for 10 seconds, followed       shown). Hematologic analysis of Bcrp1;Mdr1a/b;Mrp22/2
by centrifugation for 10 minutes at 11,300g; 100 ml of the clear          mice showed that hemoglobin levels were mildly but signifi-
supernatant was diluted in a 1:1 ratio with 100 ml of 3.5 mM              cantly reduced compared with wild-type mice (males: 6.7 6
ammonium formate buffer, pH 3.5. Urine was 10 diluted by
                                                                          0.4 mM versus 7.4 6 0.1 mM; females: 6.8 6 0.1 mM versus
addition of a 180 ml of internal standard solution in methanol-
ammonium formate buffer, pH 3.5 (30:70, v/v) to a 20-ml urine
                                                                          7.6 6 0.1 mM, n 5 3–5; P , 0.05). This was previously also found
sample aliquot.                                                           in other Mrp2-deficient mouse strains (Vlaming et al., 2006,
   LC-MS/MS Analysis of PhIP and Its Metabolites. LC-MS/MS                2009a). Other measured parameters did not show consistent
was used for the separation and detection of PhIP and its metabolites.    differences with wild-type mice (not shown). Overall, despite the
Mobile phase A was prepared by adjusting a 5 mM ammonium                  complete absence of the three main apical detoxifying ABC
formate solution to pH 3.5 with a 98% formic acid solution. Mobile        transporters, the Bcrp1;Mdr1a/b;Mrp22/2 mice displayed only
phase B consisted of methanol. Mobile phase A and B were pumped           a few modest phenotypic alterations that were previously seen
through a Synergi Hydro 110 Å column (150  2.0 mm I.D., 4 mm;            in the single knockout strains. They appear, therefore, fully
Phenomenex, Torrance, CA) at a flow rate of 0.2 ml/min.                   amenable to pharmacological and toxicological studies.
   The LC eluate was directed to an API 3000 triple quadrupole MS
                                                                             Expression Levels of Metabolizing Enzymes and ABC
(AB Sciex, Foster City, CA) equipped with an electrospray ion source
operating in the positive ion mode. For quantification, multiple-
                                                                          Transporters in Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;
reaction monitoring chromatograms were acquired and processed.            Mrp2;Mrp32/2 Mice. We determined the RNA expression
Calibration curves of analyte/internal standard peak area ratio           levels of a set of genes involved in xenobiotic metabolism and
versus, respectively, PhIP and N2-OH-PhIP concentrations were             transport in general (Cyp3a11, Cyp3a13, Cyp3a25, Cyp3a41,
constructed, and a weighted 1/x2 (the reciprocal of the squared           and Mrp4), as well as a number of genes that have been
concentration) linear regression was applied to the data. Quantifica-     associated with transport or metabolism of PhIP and/or its
tion of metabolites was performed based on the calibration curve          main metabolites (Cyp1a1, Cyp1a2, Ugt1a1, Sult1a1, Nat1,
of PhIP. PhIP-glucuronides and OH-PhIP-glucuronide metabolites            Nat2, Mdr1a, Mdr1b, and Mrp3) in liver, small intestine, and
(Supplemental Fig. 1) could not be unequivocally distinguished, as        kidney of male mice.
reference standards were not available in all cases. Therefore, in this
                                                                             In liver, expression was significantly increased for
study, PhIP-glucuronides stands for the sum of PhIP-N2-glucuronide
and PhIP-N3-glucuronide, and OH-PhIP-glucuronides is defined as
                                                                          Cyp3a11, Cyp3a13, Ugt1a1, Mrp3, and Mrp4. Cyp3a11 and
the sum of all OH-PhIP-glucuronides detected (Supplemental Fig. 1).       Cyp3a25 were mildly (2.5- and 1.5-fold, respectively) in-
After purification from mouse urine, the identity of both PhIP-49-        creased in livers of Bcrp1;Mdr1a/b;Mrp22/2 mice (Supple-
sulfate and PhIP-5-sulfate was confirmed by NMR analysis (Teunissen       mental Table 1). Cyp3a13 was increased 3.1-fold in Bcrp1;
et al., 2011). The lower limit of quantification (LLQ) of PhIP and its    Mrp2;Mrp32/2 mice (P 5 0.043), but in all strains, Cyp3a13
metabolites was 1 ng/ml.                                                  RNA levels in liver were extremely low. Mrp4 protein was
PhIP PK in Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;Mrp32/2 Mice         523
previously undetectable by Western blot in livers of wild-type     1999; Frandsen and Alexander, 2000; Gooderham et al., 2001,
FVB and Bcrp1;Mrp2;Mrp32/2 mice (Vlaming et al., 2009b).           2002; Lauber et al., 2004; Nakagama et al., 2005; Chen et al.,
Accordingly, we found only extremely low levels of Mrp4 RNA        2007; Lauber and Gooderham, 2007, 2011). 5-OH-PhIP and
in livers of all strains, even though they were increased by 28-   PhIP-5-sulfate are surrogate markers for the levels of the
and 43-fold in Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;               ultimate genotoxic nitrenium radical cation that forms DNA
Mrp32/2 mice compared with wild-type mice, respectively            adducts (Supplemental Fig. 1) (Alexander et al., 2002).
(P , 0.001 for both strains). Of the enzymes potentially              PhIP and its main metabolite, N2-OH-PhIP, were detected
involved in PhIP metabolism, Ugt1a1 mRNA was mildly                in the plasma of all mouse strains 30 minutes after
increased (1.5- and 1.6-fold, respectively; P , 0.05 for both      intravenous or oral administration of PhIP. No significant
strains) in livers of Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;             differences between the strains were found (Fig. 1A; Supple-
Mrp2;Mrp32/2 mice. Furthermore, Mrp3 mRNA levels in                mental Tables 2 and 3). After intravenous (but not oral)
Bcrp1;Mdr1a/b;Mrp22/2 mice were increased 4.5-fold com-            administration, PhIP-5-sulfate was detected in plasma of both
pared with wild-type (P 5 0.012). Moderately increased Mrp3        knockout strains but not in wild-type plasma, suggesting that
protein levels in liver were previously also found for other       Bcrp1 and/or Mrp2 influence the plasma levels of this marker
Mrp2-deficient strains (Vlaming et al., 2006, 2009a).              for genotoxic exposure (Supplemental Table 2). After oral, but
   In small intestine, only Cyp1a1 RNA was significantly           not after intravenous administration, the detoxification
reduced in Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;Mrp32/2            product 49-OH-PhIP was quite abundant in plasma of all

                                                                                                                                     Downloaded from molpharm.aspetjournals.org at ASPET Journals on February 20, 2015
mice, to 15 or 21% of wild-type values, respectively. None of      strains, but no differences between strains were found
the other tested genes displayed significant or meaningful         (Supplemental Table 3). The absence of 49-OH-PhIP in plasma
alterations in expression in small intestine (Supplemental         after intravenous administration suggests that this com-
Table 1).                                                          pound is formed primarily by intestinal enzymes shortly after
   In kidney, Cyp1a2 RNA was upregulated in Bcrp1;Mdr1a/           oral administration. Indeed, Cyp1a1 and 1a2, which can form
b;Mrp22/2, but not Bcrp1;Mrp2;Mrp32/2 mice, from an                this metabolite, are abundantly expressed in murine small
undetectable expression level in wild-type mice (.40 poly-         intestine (Ito et al., 2007). Although Cyp1a1 expression was
merase chain reaction cycles). Cyp3a11 was also virtually          reduced in small intestine of both knockout strains (Supple-
undetectable in wild-type and Bcrp1;Mrp2;Mrp32/2 kidneys           mental Table 1), apparently this did not significantly affect
and increased 64-fold in Bcrp1;Mdr1a/b;Mrp22/2 kidneys,            the formation of 49-OH-PhIP in these strains.
but it was still very low compared with expression levels in          Thirty minutes after intravenous administration, in livers
liver tissue (Supplemental Table 1). None of the other tested      of all strains, 1.8–2.5% of the given dose was found as
genes was significantly changed, with the exception of a           unchanged PhIP (Fig. 1B). Similar amounts of PhIP (2.7–
12-fold increase in Mrp4 RNA (P , 0.05) in Bcrp1;Mdr1a/b;          4.5%) were found in liver after oral administration (Fig. 1B).
Mrp22/2 kidneys, in line with what was found before for            No significant differences between the three strains were
Mrp4 protein levels in kidney of other Mrp2-deficient strains      found. Interestingly, in Bcrp1;Mrp2;Mrp32/2 mice, after both
(Vlaming et al., 2006, 2008, 2009a). Overall, changes in gene      intravenous and oral administration, the precarcinogen N2-
expression in these strains were remarkably limited given          OH-PhIP accumulated significantly in the liver (Fig. 1, C and
their genetic deficiencies in major detoxifying transporters.      D). This was not the case in the wild-type and Bcrp1;Mdr1a/b;
   Effect of ABC Transporters on Plasma and Tissue                 Mrp22/2 mice, suggesting that Mrp3 protein limits accumu-
Distribution of PhIP and Its Metabolites. As PhIP is               lation of this metabolite in the liver. After intravenous
heavily metabolized in the body (Supplemental Fig. 1), and, as     administration, the same was found for the mutagenesis
only some of the known PhIP metabolites are potentially            marker PhIP-5-sulfate (Fig. 1C). PhIP-5-sulfate was not
carcinogenic (Schut and Snyderwine, 1999; Frandsen and             detected in the liver (and plasma; see above) after oral
Alexander, 2000; Gooderham et al., 2001, 2002; Lauber et al.,      administration.
2004; Nakagama et al., 2005; Chen et al., 2007; Lauber and            The levels of the parent compound PhIP in the intestinal
Gooderham, 2007, 2011), we developed an LC-MS/MS assay             tract (contents and tissue combined) were determined after
for the quantification of PhIP and N2-OH-PhIP in mouse             intravenous (Fig. 2A) or oral (Fig. 2B) application of PhIP
matrices (Teunissen et al., 2010) and expanded this assay for      (1 mg/kg). For both administration routes, the small intestinal
detection of various phase 1 and phase 2 metabolites of PhIP       levels of PhIP in Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;
(Teunissen et al., 2011) We used this method to determine the      Mrp32/2 mice were reduced 4- to 6-fold compared with wild-
plasma and tissue concentrations of PhIP and its carcinogenic      type mice. This finding suggests an important role for
metabolites 30 minutes after oral or intravenous administra-       Bcrp1 and/or Mrp2 in the hepatobiliary and/or intestinal
tion of PhIP (1 mg/kg) to wild-type, Bcrp1;Mdr1a/b;Mrp22/2         excretion of PhIP or, for oral administration, in reducing net
and Bcrp1;Mrp2;Mrp32/2 mice. Overviews of all metabolites          intestinal uptake. As there was no difference between the
detected in the tissues of the three strains are presented         Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;Mrp32/2 mice,
in Supplemental Table 2 (intravenous administration) and           Mdr1a/b and Mrp3 are probably not involved here. Also, in
Supplemental Table 3 (oral administration). Most metabo-           cecum and colon (the latter for intravenous application only)
lites, for example all glucuronide conjugates, are considered      of both combination knockout strains, PhIP concentrations
detoxifying metabolites and do not have carcinogenic poten-        were reduced (Fig. 2, A and B). Small intestine levels of the
tial. Also, the formation of 49-OH-PhIP and subsequently           carcinogenic PhIP metabolite N2-OH-PhIP and the genotoxic
PhIP-49-sulfate (Supplemental Fig. 1) is considered a de-          marker PhIP-5-sulfate were significantly reduced compared
toxification pathway. On the other hand, besides PhIP itself,      with levels in wild-type in Bcrp1;Mdr1a/b;Mrp22/2 and
N2-OH-PhIP and the short-lived species PhIP-N-sulfate and          Bcrp1;Mrp2;Mrp32/2 mice, both after intravenous and oral
N-acetyl-PhIP are (pre)carcinogenic (Schut and Snyderwine,         administration of PhIP (Fig. 2, C and D). This result suggests
524         Vlaming et al.

                                                                                                                                                                Downloaded from molpharm.aspetjournals.org at ASPET Journals on February 20, 2015
Fig. 1. Levels of PhIP and its primary carcinogenic metabolites in plasma and liver of male wild-type, Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;
Mrp32/2 mice 30 minutes after intravenous (n = 4–11) or oral (n = 5) administration of PhIP (1 mg/kg). (A) PhIP and N2-OH-PhIP levels in plasma of the
strains after intravenous or oral administration. (B) PhIP levels in liver of the strains after intravenous or oral administration. (C) Levels of N2-OH-PhIP
and PhIP-5-sulfate in liver of the strains after intravenous administration. nd, not detected. Detection limits for N2-OH-PhIP and PhIP-5-sulfate were
0.02% and 0.01% of the dose, respectively (indicated by dashed lines). (D) N2-OH-PhIP levels in liver of the strains after oral administration. Data are
means 6 S.D. (n = 4–11; **P , 0.01; ***P , 0.001, compared with wild-type). In case tissue levels in wild-type mice were below the detection limit, the
LLQ 6 LLQ was used to calculate statistical significance.

that Bcrp1 and/or Mrp2 are involved in biliary and/or direct                   levels in small intestine of Bcrp1;Mdr1a/b;Mrp22/2 mice
intestinal excretion of these metabolites. In addition, the                    were even lower than levels in the Bcrp1;Mrp2;Mrp32/2 mice,
reduced levels of Cyp1a1 in small intestine tissue of the                      despite an insignificant difference in Cyp1a1 expression
knockout strains (Supplemental Table 1) may also lead to                       between these strains. Besides PhIP, N2-OH-PhIP, and
reduced formation of N2-OH-PhIP in small intestine. How-                       PhIP-5-sulfate, various other PhIP metabolites (glucuronide
ever, after intravenous and oral administration, N2-OH-PhIP                    and sulfate conjugates) were detected in wild-type small

                                                                                                         Fig. 2. PhIP, N2-OH-PhIP, and PhIP-5-sulfate
                                                                                                         levels in the intestinal tract (tissue and contents)
                                                                                                         of male wild-type, Bcrp1;Mdr1a/b;Mrp22/2 and
                                                                                                         Bcrp1;Mrp2;Mrp32/2 mice 30 minutes after in-
                                                                                                         travenous (n = 4–11) or oral (n = 5) administra-
                                                                                                         tion of PhIP (1 mg/kg). (A) Parent PhIP levels in
                                                                                                         the intestinal tract of the strains after intrave-
                                                                                                         nous administration. (B) Parent PhIP levels in
                                                                                                         the intestinal tract of the strains after oral
                                                                                                         administration. (C) N2-OH-PhIP levels in the
                                                                                                         small intestine of the strains after intravenous or
                                                                                                         oral administration. N2-OH-PhIP detection limit
                                                                                                         was 0.02% of the dose, as indicated by the dashed
                                                                                                         line. (D) PhIP-5-sulfate levels in the small
                                                                                                         intestine of the strains after intravenous or oral
                                                                                                         PhIP administration. PhIP-5-sulfate detection
                                                                                                         limit was 0.02% of the dose (dashed line). Data
                                                                                                         are means 6 S.D. (n = 4–11; *P , 0.05; **P ,
                                                                                                         0.01; ***P , 0.001); nd, not detected; Sm. Int.,
                                                                                                         small intestine. In cases where values were
                                                                                                         below detection levels (nd), the statistical signif-
                                                                                                         icance was calculated assuming values of LLQ 6
                                                                                                         LLQ (mean 6 S.D.).
PhIP PK in Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;Mrp32/2 Mice                        525
intestine, after both intravenous and oral administration                      after oral administration, and they were likewise increased in
(Supplemental Tables 2 and 3), but these were not detectable                   both knockout strains (Fig. 3B). Thus, the combined absence
in small intestines of the Bcrp1;Mdr1a/b;Mrp22/2 and                           of Bcrp1 and Mrp2 leads to the accumulation of carcinogenic
Bcrp1;Mrp2;Mrp32/2 mice, suggesting roles for Bcrp1 and/                       PhIP metabolites in the kidney. These increases may be due
or Mrp2 in the hepatobiliary and/or intestinal elimination of                  partly to reduced elimination of these compounds from the
these metabolites as well.                                                     kidney, but for PhIP-5-sulfate, they may also reflect higher
  PhIP levels in kidney tissue 30 minutes after intravenous or                 plasma exposure levels (Supplemental Table 2).
oral administration were not significantly different between                      As previously shown in Bcrp12/2 mice (Enokizono et al.,
the analyzed strains (Fig. 3A). However, in the kidneys of the                 2008), the brain and testis penetration of PhIP after
combination knockout mice, a significantly increased accu-                     intravenous administration was significantly increased
mulation was seen of various PhIP-metabolites, including the                   in Bcrp1-deficient mice (Fig. 3D; Supplemental Table 2). After
(pre)carcinogen N2-OH-PhIP and genotoxicity marker PhIP-                       oral administration, similar effects of Bcrp1 absence were
5-sulfate (Fig. 3, B and C). For N2-OH-PhIP after intravenous                  observed, although not statistically significant in all cases, as
(but not oral) administration of PhIP, the increased kidney                    a result of high interindividual variation (Supplemental
levels were more pronounced in the Bcrp1;Mdr1a/b;Mrp22/2                       Table 3). Interestingly, after intravenous administration,
mice than in the Bcrp1;Mrp2;Mrp32/2 mice. This may                             the brain levels of PhIP were higher in the Bcrp1;Mdr1a/b;
perhaps be caused in part by the increased expression of                       Mrp22/2 mice compared with Bcrp1;Mrp2;Mrp32/2 mice

                                                                                                                                                              Downloaded from molpharm.aspetjournals.org at ASPET Journals on February 20, 2015
Cyp1a2 in the kidney of Bcrp1;Mdr1a/b;Mrp22/2 mice,                            (0.13% 6 0.03% of the dose in Bcrp1;Mdr1a/b;Mrp22/2 mice
compared with wild-type and Bcrp1;Mrp2;Mrp32/2 mice                            versus 0.06% 6 0.03% of the dose in Bcrp1;Mrp2;Mrp32/2
(Supplemental Table 1). Note that N2-OH-PhIP plasma levels                     mice, n 5 4–7; P 5 0.007; Fig. 3D), even though PhIP plasma
were not significantly different between these strains (Fig.                   levels were similar between the strains (Fig. 1A). This finding
1A). Small amounts of PhIP-49-sulfate and PhIP-glucuronides                    suggests that besides Bcrp1, Mdr1a/b may also be involved in
were also detected in the kidney after intravenous but not                     restricting PhIP penetration into the brain. A similar effect

Fig. 3. PhIP and PhIP metabolite levels in kidney, brain, and testis of male wild-type, Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;Mrp32/2 mice 30
minutes after intravenous (n = 4–11) or oral (n = 5) administration of PhIP (1 mg/kg). (A) Parent PhIP levels in kidney of the strains after intravenous or
oral administration. (B) PhIP metabolite levels in kidney of the strains after intravenous administration. nd, not detected (below 0.003% of the dose,
indicated by the dashed line). (C) PhIP metabolite levels in kidney of the strains after oral administration. (D) PhIP levels in brain and testis of the
strains after intravenous administration. Data are means 6 S.D. (*P , 0.05; **P , 0.01; ***P , 0.001).
526       Vlaming et al.

was observed for the testis, although the difference between          the Bcrp1;Mdr1a/b;Mrp22/2 mice than in wild-type mice
Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;Mrp32/2 mice was                 (0.67 × 1023 6 1.5 × 1023% of the dose in knockout versus 14 ×
(just) not statistically significant (Fig. 3D).                       1023 6 5 × 1023% of the dose in wild-type, P 5 6.3 × 1023), even
   Effect of Apical ABC Transporters on Biliary Excre-                while the plasma concentration was markedly higher (Fig. 4,
tion of PhIP and Its Metabolites. Because the concen-                 C and D). For the other metabolites that were detected in the
trations of PhIP and many of its metabolites in the intestinal        intestinal contents (N2-OH-PhIP and OH-PhIP-glucuronides),
tract after intravenous and oral administration were signif-          no differences between wild-type and knockout mice were
icantly reduced in mice deficient in Bcrp1 and Mrp2, we               found.
hypothesized that this could be caused by reduced biliary                Effect of ABC Transporters on Urinary and Fecal
excretion of these compounds. To investigate this, we                 Excretion of PhIP and Its Metabolites. As Bcrp1 and
performed gallbladder cannulations in male wild-type and              Mrp2 apparently are the main transporters for the biliary
Bcrp1;Mdr1a/b;Mrp22/2 mice and analyzed the biliary                   excretion of PhIP and some of its metabolites, we investigated
excretion of PhIP and its metabolites in the first 60 minutes         the urinary and fecal excretion of PhIP and its metabolites in
after intravenous administration of PhIP (1 mg/kg). The               the first 24 hours after intravenous administration of PhIP
cumulative biliary excretion of PhIP was 41-fold lower in             at 1 mg/kg to wild-type, Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;
combination knockout compared with wild-type mice (Fig.               Mrp2;Mrp32/2 mice. As shown in Fig. 5A, the urinary
4A), showing that Bcrp1, Mrp2, and (possibly, to a minor              excretion of PhIP was not altered in the combination knockout

                                                                                                                                           Downloaded from molpharm.aspetjournals.org at ASPET Journals on February 20, 2015
extent) Mdr1a/b are the main transporters for biliary                 mice, consistent with the absence of differences in plasma
excretion of PhIP. The biliary excretion of PhIP metabolites          concentration of PhIP observed 30 minutes after intravenous
(sulfate and glucuronide conjugates) was also dramatically            (and oral) administration (Supplemental Tables 2 and 3). This
reduced in the Bcrp1;Mdr1a/b;Mrp2 2/2 mice (Fig. 4B).                 suggests that these three transporters are not involved in
Surprisingly, whereas the metabolite N2-OH-PhIP was                   urinary PhIP excretion (Fig. 5A). However, in line with the
clearly detected in plasma and many tissues after intravenous         reduced biliary excretion of PhIP in Bcrp1;Mdr1a/b;Mrp22/2
and oral administration (Figs. 1–3), in bile of the mice this         mice (Fig. 4A), the fecal excretion of PhIP was dramatically
compound could not be detected. This finding suggests that, in        reduced in Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;
contrast to PhIP, for this metabolite, which is formed both           Mrp32/2 mice, to 5% and 12% of wild-type excretion levels,
intrahepatically and extrahepatically (Frandsen and Alexan-           respectively (Fig. 5A).
der, 2000; Ma et al., 2007), biliary excretion is not a significant      Although urinary excretion of PhIP was not altered in the
route of elimination.                                                 combination knockout mice, the urinary excretion of some
   The plasma levels of PhIP and its metabolites at the end of        important metabolites was altered (Fig. 5B). The urinary
the gallbladder cannulation experiment are shown in Fig. 4C.          excretion of the genotoxic exposure marker PhIP-5-sulfate
PhIP concentration was not significantly altered, but signif-         and the (pre)carcinogen N2-OH-PhIP, as well as the de-
icant increases were seen in the concentrations of the                toxification products PhIP-49-sulfate and OH-PhIP-glucuronide,
genotoxicity marker PhIP-5-sulfate (5.5-fold) and of OH-              was substantially (3- to 10-fold) increased in both knockout
PhIP-glucuronide (3.3-fold) in plasma of Bcrp1;Mdr1a/b;               strains (Fig. 5B). This is in line with increased kidney
Mrp22/2 mice, possibly as a consequence of the reduced                accumulation of these compounds (Fig. 3, B and C) and may
biliary excretion of these compounds. N2-OH-PhIP levels in            well reflect overall higher systemic exposure levels. No
plasma were relatively low and not significantly different            significant differences were seen in urinary excretion of
between the strains (Fig. 4C). Interestingly, in the livers of        PhIP and its metabolites between Bcrp1;Mdr1a/b;Mrp22/2
Bcrp1;Mdr1a/b;Mrp22/2 mice, despite highly decreased                  and Bcrp1;Mrp2;Mrp32/2 mice, suggesting a primary role
biliary excretion (Fig. 4B) and mildly increased liver RNA            for Bcrp1 and Mrp2. Because of the abundant presence of
levels of Ugt1a1, OH-PhIP-glucuronide levels were signifi-            many interfering compounds, PhIP metabolite concentra-
cantly reduced (0.06% 6 0.02% of the dose in knockout versus          tions in the feces of the mice could not be reliably quantified.
0.14% 6 0.05% of the dose in wild-type, P 5 5.2 × 1023). This
result suggests increased liver elimination of OH-PhIP-
glucuronides over the sinusoidal membrane in the Bcrp1;
                                                                                               Discussion
Mdr1a/b;Mrp22/2 mice. PhIP levels in the liver of these                  We show here that Bcrp1;Mdr1a/b;Mrp22/2 mice, which
mice, like in plasma (Fig. 4C), tended to be somewhat higher          lack all the major apical multidrug efflux transporters, are
in the knockout strain, but this difference was not significant       viable and fertile and show no obvious phenotypic aberrations
(4.4% 6 2.1% of the dose in knockout versus 2.9% 6 1.0% of            other than an increased liver weight and increased plasma
the dose in wild-type, P 5 0.18). Other PhIP metabolites were         bilirubin levels, as previously found in Bcrp1;Mrp22/2 mice
not detected in livers of the mice after the gallbladder              (Vlaming et al., 2009a). In addition, only a few minor or modest
cannulation experiment. To investigate the effect of Bcrp1,           changes in RNA expression of some drug transporters and
Mrp2 and/or Mdr1a/b on the direct intestinal secretion of             metabolizing enzymes were observed. These mice should
PhIP and its metabolites, we also analyzed the small                  therefore be valuable tools for studies on the relative and
intestinal contents of the mice after gall bladder cannulations       combined effects of Bcrp1, Mdr1a/b and Mrp2 on the
(Fig. 4D). The amount of PhIP in the small intestinal contents        pharmacokinetics, toxicity, and carcinogenicity of shared
was not significantly different between wild-type and Bcrp1;          substrates in vivo, especially in combination with the pre-
Mdr1a/b;Mrp22/2 mice, suggesting that these transporters              viously generated single and double knockout mice for these
do not affect direct intestinal excretion of unchanged PhIP.          transporters (Jonker et al., 2002; Vlaming et al., 2006, 2009a,b).
However, the direct intestinal excretion of PhIP-5-sulfate               The exact mechanism behind the increase in liver weight
was, like the biliary excretion (Fig. 4B), dramatically lower in      of the two mouse strains studied here, and of other knockout
PhIP PK in Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;Mrp32/2 Mice                       527

                                                                                                                                                             Downloaded from molpharm.aspetjournals.org at ASPET Journals on February 20, 2015
Fig. 4. Biliary excretion and plasma and small intestinal content levels of PhIP and its metabolites in the first 60 minutes after intravenous
administration of PhIP (1 mg/kg) to male gallbladder-cannulated wild-type and Bcrp1;Mdr1a/b;Mrp22/2 mice. (A) Cumulative biliary excretion of
parent PhIP. (B) Cumulative biliary excretion of PhIP metabolites. PhIP-5-sulfate was not detected (nd) in Bcrp1;Mdr1a/b;Mrp22/2 bile (LLQ was
1 × 1024% of the dose, indicated by the dashed line). N2-OH-PhIP and PhIP-49-sulfate were not detectable in wild-type or knockout bile. (C) Plasma
concentrations of PhIP and its metabolites at t = 60 minutes. (D) Small intestinal contents (Sm. Int. Cont.) levels of PhIP and PhIP metabolites at t = 60
minutes. Data are means 6 S.D. (n = 5; *P , 0.05; **P , 0.01; ***P , 0.001). In cases where values were below detection levels (nd), the statistical
significance was calculated assuming values of LLQ 6 LLQ (mean 6 S.D.).

strains lacking Mrp2 (Vlaming et al., 2006, 2009a,b), is not                  hours was markedly reduced. Furthermore, the combined
known. Likely, reduced detoxification of some compound(s)                     absence of Bcrp1 and Mrp2 leads to increased exposure of
that can affect liver size causes this effect, but the modest                 plasma, liver, and kidney to potentially carcinogenic PhIP
changes we observed in expression of several other                            metabolites like N2-OH-PhIP, as well as PhIP-5-sulfate, and
functional detoxifying systems do not suggest a drastic                       increased urinary excretion of all of these compounds. As most
alteration in the overall functioning of the liver in these                   PhIP-metabolizing enzymes, except for the mildly increased
mice.                                                                         Ugt1a1 in liver, were not differently expressed in both strains
  In this first study with Bcrp1;Mdr1a/b;Mrp22/2mice,                         compared with wild-type (Supplemental Table 1), the effects
combined with the Bcrp1;Mrp2;Mrp32/2 mice (Vlaming                            observed are likely mostly caused by absence of Bcrp1 and/or
et al., 2009b), we show that Bcrp1 and Mrp2 are the main                      Mrp2. It appears that, when Bcrp1 and/or Mrp2 are absent,
transporters responsible for the biliary, intestinal and fecal                PhIP is less readily removed from the body and probably
excretion of the dietary carcinogen PhIP, as well as for the                  more extensively converted to metabolites, including carci-
biliary and/or direct intestinal excretion of the genotoxic                   nogenic ones. Since they also affect tissue distribution and
exposure marker PhIP-5-sulfate and several other PhIP                         elimination of these metabolites (Figs. 2–5), Bcrp1 and Mrp2
metabolites. The urinary excretion of PhIP was not altered                    may well be involved in protecting the body from PhIP-
in the knockout mice, but the total PhIP excretion over 24                    induced carcinogenesis.
528         Vlaming et al.

                                                                                                                                                              Downloaded from molpharm.aspetjournals.org at ASPET Journals on February 20, 2015
Fig. 5. Urinary and fecal excretion of PhIP and its metabolites in the first 24 hours after intravenous administration of PhIP (1 mg/kg) to male wild-type,
Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;Mrp32/2 mice. (A) Urinary and fecal excretion of parent PhIP in the different strains. (B) Urinary excretion
of PhIP metabolites in the different strains. Data are means 6 S.D. (n = 5; *P , 0.05; **P , 0.01; ***P , 0.001). PhIP metabolite concentrations in the
feces could not be reliably quantified because of the presence of interfering compounds.

   We additionally found that Mrp3 limits exposure of the                      et al., 2012), this indicates that the combined deficiency of
liver to N2-OH-PhIP and PhIP-5-sulfate, presumably by                          Bcrp1 and Mdr1a/1b results in higher brain accumulation of
mediating elimination of these compounds, after their                          PhIP than the single deficiency of Bcrp1. Thus, not only Bcrp1
formation in the liver, across the sinusoidal membrane into                    (Enokizono et al., 2008), but also Mdr1a/1b contributes to
the blood. This process could thus have consequences for the                   protection of the brain from PhIP, consistent with the
carcinogenic potential of N2-OH-PhIP elsewhere in the body.                    observed modest in vitro transport of PhIP by Mdr1a (van
However, we found substantially increased kidney levels and                    Herwaarden et al., 2003). The same may apply for the testis
urinary excretion of N2-OH-PhIP and PhIP-5-sulfate in both                     penetration of PhIP in the combination knockout strains
knockout strains (Figs. 3, B and C, and 5B). Also the overall                  (Fig. 3D).
urinary excretion of these compounds was not different                           This study shows that Bcrp1, Mrp2, and to some extent
between Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;Mrp32/2                           Mdr1a/1b have a major impact on the exposure of the body to
mice (Fig. 5B). We therefore conclude that although Mrp3                       PhIP, both its primary carcinogenic metabolites and a geno-
seems important for short-term sinusoidal liver elimination of                 toxicity exposure marker. On the one hand, Bcrp1 and Mrp2
N2-OH-PhIP and PhIP-5-sulfate, this does not seem to                           deficiency resulted in markedly reduced biliary and direct
markedly affect their systemic exposure levels. Possibly other                 intestinal excretion of PhIP and PhIP-5-sulfate and markedly
basolateral ABC transporters, such as Mrp4, can also gradually                 reduced short-term intestinal exposure to PhIP, N2-OH-PhIP,
transport these compounds from the liver (albeit more slowly                   and PhIP-5-sulfate as well as strongly decreased fecal
than Mrp3), or other tissues may contribute to N2-OH-PhIP                      excretion of PhIP. On the other hand, the same deficiency
and PhIP-5-sulfate formation as well.                                          resulted in higher kidney and urinary levels of N2-OH-PhIP,
   Similar to studies on rat Mrp2 by Dietrich et al. (2001a), we               and markedly higher plasma, kidney, and urinary levels of
found that mouse Bcrp1, Mrp2, and Mdr1a/b are not essential                    PhIP-5-sulfate. These data further indicate that the overall
for the urinary excretion of PhIP and its metabolites. Urinary                 systemic exposure to carcinogenic PhIP metabolites is in-
excretion of PhIP in each of the Bcrp1;Mrp2-deficient strains                  creased, whereas the intestinal exposure is decreased by
was comparable to that in wild-type mice, and for many PhIP                    Bcrp1/Mrp2 deficiency. It is therefore possible that Bcrp1/
metabolites, urinary excretion was even increased in the                       Mrp2 deficiency would have a pronounced effect on the
knockouts, most likely as a consequence of increased plasma                    number and tissue distribution of PhIP-induced tumors, but
concentrations of these compounds (Fig. 5). The latter may                     the direction of such effects in various tissues (increased or
result from reduced elimination by ABC transporters and                        decreased tumor formation) may not be so easy to predict.
occasionally from mild increases in the formation of metab-                    Ultimately, only in vivo carcinogenesis studies can address
olites, for instance, from upregulation of Ugt1a1 in the liver of              these questions.
both knockout strains and of Cyp1a2 in kidney tissue of                          It is interesting to note that Mrp2 (apical) and Mrp3
Bcrp1;Mdr1a/b;Mrp22/2 mice (Supplemental Table 1).                             (basolateral) are generally expressed on opposite poles of
   It is interesting to note that the brain accumulation of PhIP               polarized cells in tissues analyzed in this study, such as
was significantly, if modestly, increased in the Bcrp1;Mdr1a/                  hepatocytes, enterocytes, and kidney tubular epithelial cells.
b;Mrp22/2 mice compared with Bcrp1;Mrp2;Mrp32/2 mice                           This might result in strongly increased tissue accumulation of
(Fig. 3D). Since Mrp2 and Mrp3 are unlikely to play a role in                  shared substrates of both of these transporters, when both are
the blood-brain barrier of FVB mice, where they are not                        deficient (i.e., in Bcrp1;Mrp2;Mrp32/2 mice). Survey of our
detectably expressed (Soontornmalai et al., 2006; Agarwal                      data, however, shows only a few metabolites that are consistently
PhIP PK in Bcrp1;Mdr1a/b;Mrp22/2 and Bcrp1;Mrp2;Mrp32/2 Mice                                  529
and significantly accumulating more strongly in Bcrp1;Mrp2;                                Frandsen H and Alexander J (2000) N-acetyltransferase-dependent activation of
                                                                                             2-hydroxyamino-1-methyl-6-phenylimidazo[4,5-b]pyridine: formation of 2-amino-1-
Mrp32/2 than in Bcrp1;Mdr1a/b;Mrp22/2 tissues. These include                                 methyl-6-(5-hydroxy)phenylimidazo [4,5-b]pyridine, a possible biomarker for the
N2-OH-PhIP and PhIP-5-sulfate in liver, N2-OH-PhIP in small                                  reactive dose of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine. Carcinogenesis
                                                                                             21:1197–1203.
intestine and PhIP-4’-sulfate in kidney. Given the substantial                             Gooderham NJ, Murray S, Lynch AM, Yadollahi-Farsani M, Zhao K, Boobis AR,
other alterations in general PhIP and PhIP metabolite                                        and Davies DS (2001) Food-derived heterocyclic amine mutagens: variable me-
                                                                                             tabolism and significance to humans. Drug Metab Dispos 29:529–534.
disposition in these mice, we think such shifts may be difficult                           Gooderham NJ, Zhu H, Lauber S, Boyce A, and Creton S (2002) Molecular and
to interpret in a straightforward manner.                                                    genetic toxicology of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP).
                                                                                             Mutat Res 506-507:91–99.
  Many polymorphisms and mutations in BCRP, MRP2,                                          Huang Y (2007) Pharmacogenetics/genomics of membrane transporters in cancer
MRP3, and P-gp are known, and these often lead to reduced                                    chemotherapy. Cancer Metastasis Rev 26:183–201.
function (Huang, 2007; Maeda and Sugiyama, 2008). There                                    Ito S, Chen C, Satoh J, Yim S, and Gonzalez FJ (2007) Dietary phytochemicals
                                                                                             regulate whole-body CYP1A1 expression through an arylhydrocarbon receptor
are even substantial numbers of individuals with partial or                                  nuclear translocator-dependent system in gut. J Clin Invest 117:1940–1950.
complete genetic deficiencies in BCRP [e.g., Jr(a-) individuals]                           Jonker JW, Buitelaar M, Wagenaar E, Van Der Valk MA, Scheffer GL, Scheper RJ,
                                                                                             Plosch T, Kuipers F, Elferink RP, and Rosing H et al. (2002) The breast cancer
(Saison et al., 2012) or in MRP2 (Dubin-Johnson syndrome).                                   resistance protein protects against a major chlorophyll-derived dietary phototoxin
Since PhIP is an abundant carcinogen, it will be of interest                                 and protoporphyria. Proc Natl Acad Sci USA 99:15649–15654.
                                                                                           Lauber SN, Ali S, and Gooderham NJ (2004) The cooked food derived carcinogen
to assess the effect of such ABC transporter deficiencies on                                 2-amino-1-methyl-6-phenylimidazo[4,5-b] pyridine is a potent oestrogen: a mech-
the carcinogenic potential of PhIP in epidemiologic studies.                                 anistic basis for its tissue-specific carcinogenicity. Carcinogenesis 25:
                                                                                             2509–2517.
Moreover, one can reasonably predict that the disposition of

                                                                                                                                                                                    Downloaded from molpharm.aspetjournals.org at ASPET Journals on February 20, 2015
                                                                                           Lauber SN and Gooderham NJ (2007) The cooked meat derived genotoxic carcinogen
many other dietary and environmental carcinogens, and their                                  2-amino-3-methylimidazo[4,5-b]pyridine has potent hormone-like activity: mecha-
                                                                                             nistic support for a role in breast cancer. Cancer Res 67:9597–9602.
hydroxylated, and sulfate- and glucuronide-conjugated acti-                                Lauber SN and Gooderham NJ (2011) The cooked meat-derived mammary carcino-
vated derivatives will be markedly affected by these ABC                                     gen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine promotes invasive behav-
transporters as well. Our newly generated Bcrp1;Mdr1a/b;                                     iour of breast cancer cells. Toxicology 279:139–145.
                                                                                           Leslie EM, Deeley RG, and Cole SP (2005) Multidrug resistance proteins: role of
Mrp22/2 and Bcrp1;Mrp2;Mrp32/2 mice should therefore not                                     P-glycoprotein, MRP1, MRP2, and BCRP (ABCG2) in tissue defense. Toxicol Appl
only be valuable tools for pharmacokinetic studies of drugs,                                 Pharmacol 204:216–237.
                                                                                           Ma X, Idle JR, Malfatti MA, Krausz KW, Nebert DW, Chen CS, Felton JS, Waxman
but also for studies on the effect of the different ABC transporters                         DJ, and Gonzalez FJ (2007) Mouse lung CYP1A1 catalyzes the metabolic activation
on limiting (or possibly increasing) xenobiotic-induced carcino-                             of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP). Carcinogenesis 28:
                                                                                             732–737.
genesis in vivo.                                                                           Maeda K and Sugiyama Y (2008) Impact of genetic polymorphisms of transporters on
                                                                                             the pharmacokinetic, pharmacodynamic and toxicological properties of anionic
Acknowledgments                                                                              drugs. Drug Metab Pharmacokinet 23:223–235.
                                                                                           Nakagama H, Nakanishi M, and Ochiai M (2005) Modeling human colon cancer in
  The authors thank their colleagues for critical reading of the                             rodents using a food-borne carcinogen, PhIP. Cancer Sci 96:627–636.
manuscript; Rob Lodewijks, Enver Delic, and Hans Tensen for                                Saison C, Helias V, Ballif BA, Peyrard T, Puy H, Miyazaki T, Perrot S, Vayssier-
                                                                                             Taussat M, Waldner M, and Le Pennec PY et al. (2012) Null alleles of ABCG2
excellent technical assistance; and Martin van der Valk and Ji-Ying                          encoding the breast cancer resistance protein define the new blood group system
Song for histological analysis.                                                              Junior. Nat Genet 44:174–177.
                                                                                           Schinkel AH, Mayer U, Wagenaar E, Mol CA, van Deemter L, Smit JJ, van der Valk
                                                                                             MA, Voordouw AC, Spits H, and van Tellingen O et al. (1997) Normal viability and
Authorship Contributions
                                                                                             altered pharmacokinetics in mice lacking mdr1-type (drug-transporting) P-glyco-
  Participated in research design: Vlaming, Teunissen, Rosing,                               proteins. Proc Natl Acad Sci USA 94:4028–4033.
Schellens, Beijnen, Schinkel.                                                              Schinkel AH and Jonker JW (2003) Mammalian drug efflux transporters of the ATP
                                                                                             binding cassette (ABC) family: an overview. Adv Drug Deliv Rev 55:3–29.
  Conducted experiments: Vlaming, Teunissen, van de Steeg, van                             Schut HA and Snyderwine EG (1999) DNA adducts of heterocyclic amine food
Esch, Wagenaar, de Greef.                                                                    mutagens: implications for mutagenesis and carcinogenesis. Carcinogenesis 20:
  Contributed new reagents or analytic tools: Teunissen, Brunsveld,                          353–368.
                                                                                           Soontornmalai A, Vlaming MLH, and Fritschy JM (2006) Differential, strain-specific
de Greef, Rosing, Beijnen.                                                                   cellular and subcellular distribution of multidrug transporters in murine choroid
  Performed data analysis: Vlaming, Teunissen, Schinkel.                                     plexus and blood-brain barrier. Neuroscience 138:159–169.
  Wrote or contributed to the writing of the manuscript: Vlaming,                          Teunissen SF, Vlaming MLH, Rosing H, Schellens JHM, Schinkel AH, and Beijnen
                                                                                             JH (2010) Development and validation of a liquid chromatography-tandem mass
Teunissen, Schinkel.                                                                         spectrometry assay for the analysis of 2-amino-1-methyl-6-phenylimidazo[4,5-b]
                                                                                             pyridine (PhIP) and its metabolite 2-hydroxyamino-1-methyl-6-phenylimidazo[4,5-
References                                                                                   b]pyridine (N-OH-PhIP) in plasma, urine, bile, intestinal contents, faeces and eight
Agarwal S, Uchida Y, Mittapalli RK, Sane R, Terasaki T, and Elmquist WF                      selected tissues from mice. J Chromatogr B Analyt Technol Biomed Life Sci 878:
  (2012) Quantitative proteomics of transporter expression in brain capillary                2353–2362.
  endothelial cells isolated from P-glycoprotein (P-gp), breast cancer re-                 Teunissen SF, Rosing H, Brunsveld L, de Greef TFA, Durmus S, Schellens JHM,
  sistance protein (Bcrp), and P-gp/Bcrp knockout mice. Drug Metab Dispos                    Schinkel AH, and Beijnen JH (2011) Analysis of 2-Amino-1-methyl-6-
  40:1164–1169.                                                                              phenylimidazo[4,5-b]pyridine and its phase I and phase II metabolites in mouse
Alexander J, Reistad R, Hegstad S, Frandsen H, Ingebrigtsen K, Paulsen JE,                   urine using LC-UV-MS-MS. Chromatographia 74:215–226.
  and Becher G (2002) Biomarkers of exposure to heterocyclic amines: approaches to         Tian X, Swift B, Zamek-Gliszczynski MJ, Belinsky MG, Kruh GD, and Brouwer KL
  improve the exposure assessment. Food Chem Toxicol 40:1131–1137.                           (2008) Impact of basolateral multidrug resistance-associated protein (Mrp) 3 and
Borst P and Elferink RO (2002) Mammalian ABC transporters in health and disease.             Mrp4 on the hepatobiliary disposition of fexofenadine in perfused mouse livers.
  Annu Rev Biochem 71:537–592.                                                               Drug Metab Dispos 36:911–915.
Chen C, Ma X, Malfatti MA, Krausz KW, Kimura S, Felton JS, Idle JR, and Gonzalez           van Herwaarden AE, Jonker JW, Wagenaar E, Brinkhuis RF, Schellens JHM,
  FJ (2007) A comprehensive investigation of 2-amino-1-methyl-6-phenylimidazo                Beijnen JH, and Schinkel AH (2003) The breast cancer resistance protein (Bcrp1/
  [4,5-b]pyridine (PhIP) metabolism in the mouse using a multivariate data analysis          Abcg2) restricts exposure to the dietary carcinogen 2-amino-1-methyl-6-phenyl-
  approach. Chem Res Toxicol 20:531–542.                                                     imidazo[4,5-b]pyridine. Cancer Res 63:6447–6452.
Dietrich CG, de Waart DR, Ottenhoff R, Bootsma AH, van Gennip AH, and Elferink             van Herwaarden AE, Wagenaar E, Karnekamp B, Merino G, Jonker JW, and Schinkel
  RP (2001a) Mrp2-deficiency in the rat impairs biliary and intestinal excretion and         AH (2006) Breast cancer resistance protein (Bcrp1/Abcg2) reduces systemic exposure
  influences metabolism and disposition of the food-derived carcinogen 2-amino-1-            of the dietary carcinogens aflatoxin B1, IQ and Trp-P-1 but also mediates their se-
  methyl-6-phenylimidazo. Carcinogenesis 22:805–811.                                         cretion into breast milk. Carcinogenesis 27:123–130.
Dietrich CG, de Waart DR, Ottenhoff R, Schoots IG, and Elferink RP (2001b) Increased       van Waterschoot RAB, van Herwaarden AE, Lagas JS, Sparidans RW, Wagenaar E,
  bioavailability of the food-derived carcinogen 2-amino-1-methyl-6-phenylimidazo[4,5-       van der Kruijssen CM, Goldstein JA, Zeldin DC, Beijnen JH, and Schinkel
  b]pyridine in MRP2-deficient rats. Mol Pharmacol 59:974–980.                               AH (2008) Midazolam metabolism in cytochrome P450 3A knockout mice
Donner MG and Keppler D (2001) Up-regulation of basolateral multidrug resistance             can be attributed to up-regulated CYP2C enzymes. Mol Pharmacol 73:
  protein 3 (Mrp3) in cholestatic rat liver. Hepatology 34:351–359.                          1029–1036.
Enokizono J, Kusuhara H, Ose A, Schinkel AH, and Sugiyama Y (2008) Quantitative            Vlaming MLH, Mohrmann K, Wagenaar E, de Waart DR, Elferink RP, Lagas JS, van
  investigation of the role of breast cancer resistance protein (Bcrp/Abcg2) in limiting     Tellingen O, Vainchtein LD, Rosing H, and Beijnen JH et al. (2006) Carcinogen and
  brain and testis penetration of xenobiotic compounds. Drug Metab Dispos 36:                anticancer drug transport by Mrp2 in vivo: studies using Mrp2 (Abcc2) knockout
  995–1002.                                                                                  mice. J Pharmacol Exp Ther 318:319–327.
You can also read